The composition of the antigen-specific T cell receptor (TcR) repertoire is a critical determinant of the immune status of an individual. It has been shown that 'holes' in the T cell repertoire can lead to unresponsiveness to certain antigens, while escape of T cells from the process of negative selection can lead to self-reactive TcR expression and autoimmune disease. Thus, understanding the mechanisms used to generate the T cell repertoire are critical, and represent the major long term objective of this proposal. Current dogma dictates that only those T cells passing two developmental checkpoints in the thymus, positive selection and negative selection, can contribute to the mature T cell repertoire, and that these selective processes operate with the same TcR ligands (self peptides + self MHC), with relative affinity being the sole determinant of the outcome. The recent discovery of a thymus-specific component of proteasomes, the proteases that produce the self-peptides that bind to MHC molecules to create these ligands, challenges this dogma, and strongly suggests that a unique set of peptide ligands is used for positive selection. However, T cell activation by such a unique set of ligands has never been directly demonstrated, and is the major specific aim of this proposal. Specifically, the thymus-specific proteasome subunit (5t) will be ectopically expressed in non-thymic antigen presenting cells. In order to determine whether these cells express new peptide/MHC ligands that can be recognized by the naturally selected T cell repertoire, normal splenic or lymph node T cells will be mixed in vitro, and selected T cell responses will be measured, including proximal events such as changes in intracellular calcium levels and ZAP70 and erk kinase acticvation, as well as downstream events including proliferation, cytokine production and differentiation into effector cells. In vivo responses to 5t-expressing cells will be observed in adoptive-transfer and tumor transplantation models to determine whether in vitro responses correlate with in vivo recognition. To determine whether alterations in protease activity in otherwise normal host cells can lead to autoimmune responses, 5t will be expressed in transgenic animals under the control of regulatable, tissue-specific promoters. These experiments will either validate or contradict the proposed role of thymic-specific peptide ligands during T cell selection and development.